Mercurial > repos > public > sbplib_julia

comparison TODO.md @ 323:b2ddc5e4d41a

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Merge
author Vidar Stiernström <vidar.stiernstrom@it.uu.se>
date Thu, 24 Sep 2020 21:04:25 +0200
parents d705b397aa33
children 535f1bff4bcc
comparison
equal deleted inserted replaced
322:777063b6f049 323:b2ddc5e4d41a
6 6
7 ## Coding 7 ## Coding
8 - [ ] Add new Laplace opertor to DiffOps, probably named WaveEqOp(?!!?) 8 - [ ] Add new Laplace opertor to DiffOps, probably named WaveEqOp(?!!?)
9 - [ ] Add 1D operators (D1, D2, e, d ... ) as TensorOperators 9 - [ ] Add 1D operators (D1, D2, e, d ... ) as TensorOperators
10 - [ ] Create a struct that bundles the necessary Tensor operators for solving the wave equation. 10 - [ ] Create a struct that bundles the necessary Tensor operators for solving the wave equation.
11 - [ ] Use traits like IndexStyle, IndexLinear, IndexCartesian to differentiate 11 - [ ] Add a quick and simple way of running all tests for all subpackages.
12 TensorMappings that are flexible in size and those that are fixed in size
13 - [ ] Use traits for symmetric tensor mappings such that apply_transpoe = apply for all such mappings
14 - [x] Move Laplace tensor operator to different package
15 - [x] Remove grid as a property of the Laplace tensor operator
16 - [ ] Update how dependencies are handled for tests. This was updated in Julia v1.2 and would allow us to use test specific dev packages.
17 12
18 ## Reasearch and thinking
19 - [ ] Redo all Tensor applys to take Vararg instead of tuple of Index?
20 Have we been down that road before? Is there any reason not to do this?
21 - [ ] Check how the native julia doc generator works
22 - [ ] Check if Vidars design docs fit in there
23 - [ ] Formalize how range_size() and domain_size() are supposed to work in TensorMappings where dim(domain) != dim(range) (add tests or document)
24 - [x] Should there be some kind of collection struct for SBP operators (as TensorOperators), providing easy access to all parts (D2, e, d , -> YES!
25 H.. H_gamma etc.)
26 - [x] Is "missing" a good value for unknown dimension sizes (of `e*g` for example)
27 - [ ] Create a macro @lazy which replaces a binary op (+,-) by its lazy equivalent? Would be a neat way to indicate which evaluations are lazy without cluttering/confusing with special characters.
28 13
29 # Wrap up task 14 # Wrap up tasks
30
31 - [ ] Kolla att vi har @inbounds och @propagate_inbounds på rätt ställen 15 - [ ] Kolla att vi har @inbounds och @propagate_inbounds på rätt ställen
32 - [ ] Kolla att vi gör boundschecks överallt och att de är markerade med @boundscheck 16 - [ ] Kolla att vi gör boundschecks överallt och att de är markerade med @boundscheck
33 - [ ] Kolla att vi har @inline på rätt ställen 17 - [ ] Kolla att vi har @inline på rätt ställen
34 - [ ] Profilera 18 - [ ] Profilera
35 19
40 24
41 Borde man alltid skicka in N som parameter i apply_2nd_derivative, t.ex som i 25 Borde man alltid skicka in N som parameter i apply_2nd_derivative, t.ex som i
42 apply_quadrature? 26 apply_quadrature?
43 27
44 Just nu agerar apply_normal_derivative, apply_boundary_value på inte på v som 28 Just nu agerar apply_normal_derivative, apply_boundary_value på inte på v som
45 en vektor, utan randvärdet plockas ut utanför. Känns inte konsistent med övrig 29 en vektor, utan randvärdet plockas ut utanför. Känns inte konsistent med övrig design.
46 design
47
48 Specificera operatorer i TOML eller något liknande?